Influence of glycoprotein IIb/IIIa inhibitors on bleeding events after successful resuscitation and percutaneous coronary intervention

AIM: Cardiac arrest is the most serious complication in acute coronary syndromes. Glycoprotein IIb/IIIa inhibitors (GPI) are used in selected acute coronary syndrome patients. If the use of GPI leads to an increase in bleeding events and influences survival in patients after cardiac arrest is unknown. METHODS: We report retrospective data of a single center registry of patients after successful intra- and out-of-hospital cardiac arrest between 2002 and 2013. Inclusion criteria were survival for at least 6 h and successful percutaneous coronary intervention (PCI) within the first 24 h. Patients treated with other fibrinolytic agents or being supported by an extracorporeal life support system were excluded from the analysis. RESULTS: 310 patients were included in our study. 204 received GPI (GPI+), 106 did not (GPI−). Patients in the GPI+ group were significantly younger (62.8 vs. 68.0 years, p < 0.001) and had larger myocardial infarction sizes (maximum creatine kinase 3407 vs. 1450 U/l, p < 0.001). CPR duration, SOFA score and first lactate did not differ between the groups. Any bleeding occurred significantly more often in the GPI+ group (83.3% vs. 67.0%, p = 0.001). Decline of hemoglobin within the first 24 h was higher in the GPI+ group (−1.59 ± 1.71 mg/dl vs. −0.88 ± 1.95 mg/dl, p = 0.004), number of transfused packed red blood cells in the first 4 days, however, were similar (1.18 ± 0.40 vs. 0.90 ± 0.41 packs, p = 0.378). Survival at ICU discharge was significantly higher in the GPI+ group (77.5% vs. 63.2%, p = 0.008). The use of GPI was an independent predictor of hospital survival (OR 3.07, CI 1.31−7.20, p = 0.010). The positive effect for GPI persisted after nearest neighbor propensity score matching including 144 patients (OR 3.27, 95% CI 1.48−7.21, p = 0.003). CONCLUSION: After cardiac arrest, bleeding incidence was significantly higher in patients treated with GPI. Incidence of bleedings requiring transfusion, however, was similar. In this retrospective analysis, the use of GPI was an independent predictor of hospital survival. We suggest that GPI may not be withheld from cardiac arrest survivors due to potential risk of bleeding. GRAPHIC ABSTRACT: [Image: see text]